homepage Welcome to WebmasterWorld Guest from 174.129.80.166
register, free tools, login, search, subscribe, help, library, announcements, recent posts, open posts,
Subscribe to WebmasterWorld

Visit PubCon.com
Home / Forums Index / Google / Google AdWords
Forum Library, Charter, Moderators: buckworks & eWhisper & skibum

Google AdWords Forum

This 54 message thread spans 2 pages: 54 ( [1] 2 > >     
Changes in Ranking Formula
Were we notified of this?
nyet




msg:1111860
 6:03 pm on Apr 19, 2005 (gmt 0)

Continued from:
[webmasterworld.com...]



from: [webmasterworld.com ]

AWA SAID :
The difference is that now, the CTR of the ad copy itself is factored in, instead of it being solely the CTR of the keyword. Which only makes sense, IMO, given that it is the quality of the keyword and the particular ad it brings up that defines relevance, for a given search.

Was there a notification on this sent to advertisers?

To what extent is it "factored-in"?

Here is my problem. We did not know of this change. We recently (a month and a half ago) changed all the wording of all of ours ads. Under the "old" formula this would not have had an effect on our average CPC unless, over time, the new ads performed less well than the old wording.

But because our new ads don't have a history of CTR then this must be why our CPC went up 20% overnight! We thought we had a bidding war on our hands but it looks like it is the new ranking formula which increased our buy by 20%.

I am certainly glad that AWA mentioned this here some time ago, but we consider emails from Google to be offical announcements.

Why was this fundamental change NOT disclosed to us?

When I called G, i cited these posts and still the rep said AD CTR ws NOT factored in.

I am not trying to pick a fight, just trying to understand. I am also trying NOT to get angry about spending all this money we shouldn't have spent.

If I had been informed about this change I would have been much more cautious in ad rewording.

 

FromRocky




msg:1111861
 7:40 pm on Apr 19, 2005 (gmt 0)

AWA SAID :
The difference is that now, the CTR of the ad copy itself is factored in, instead of it being solely the CTR of the keyword. Which only makes sense, IMO, given that it is the quality of the keyword and the particular ad it brings up that defines relevance, for a given search.

This can't be true. Since it's true, the ad placement for a keyword in a multi-ad copy campaign will be always lower than an ad in a single ad-copy campaign. You can beat a CTR of 100%. So what's the purpose of a multi ad-copy campaign? Just to lower the ad position? The more ad-copies you write in a campaign, the lower your ad position you get.

limitup




msg:1111862
 8:23 pm on Apr 19, 2005 (gmt 0)

I have evidence of this too and I really would like to know the answer. We changed some ads recently and our CPC went up literally 50% overnight. I posted about this and got no real responses, and I received no response from G whatsover. Like you at first I thought we had a bidding war on our hands, but we know all the top advertisers and I spoke to them all and no one raised their bids. They have no reason to lie so I could only assume it was a "problem" with G. This makes sense if that is really how it works now. Pretty crappy they don't let us advertisers know ...

You can beat a CTR of 100%.

You can't have a single ad with a 100% CTR either LOL.

Adding new ads could help your CTR under this new assumed system, just not in the first day or two. But if you had 1 ad that had a 2% CTR and you added another ad that has a 5% CTR, after the new ad has at least 1000 impressions or more I would assume it would start helping you rather than hurting.

nyet




msg:1111863
 8:37 pm on Apr 19, 2005 (gmt 0)

"1st day or 2" depends on traffic. We only get about 150-250 hits per day. It has been weeks and s-l-o-w-l-y the CPC is drifting down to previoius levels.

Our CTR is HIGHER for the new ads but so is the CPC.

So far we are out a few thousands of dollars "extra". our average CPC is .98 post ad change and was .75 before.

limitup




msg:1111864
 8:58 pm on Apr 19, 2005 (gmt 0)

Yeah who knows what kind of history is needed. It could be a lot more than 1k impressions ...

irish_john




msg:1111865
 6:20 pm on Apr 27, 2005 (gmt 0)

Has anyone, noticed a recent change in the adwords algorithm? I am noticing a rough 10% increase in average cpc.

zeus661




msg:1111866
 7:15 pm on Apr 27, 2005 (gmt 0)

Can you explain "adwords algorithm" to a dummy? Thanks

nyet




msg:1111867
 7:54 pm on Apr 27, 2005 (gmt 0)

Much discussed and much noticed:

[webmasterworld.com...]

[webmasterworld.com...]

Google now factors in your ad text CTR as "ad relevance".

So near as I can tell if you change the text of your ads the average CPC will go immeadiately up until the new ad's CTR history "overcomes" the old ad.

Have you changed your ad text recently? Can you correlate it with the increase in CPC?

AdWordsAdvisor




msg:1111868
 9:51 pm on Apr 27, 2005 (gmt 0)

Has anyone, noticed a recent change in the adwords algorithm? I am noticing a rough 10% increase in average cpc.

In addition to the possibility of algo changes already noted, it could also simply be an increase in the number of competitors, or an increase in competitiveness (meaning an improved CTR, or raised Max CPC) from existing competition.

AWA

patient2all




msg:1111869
 4:05 pm on Apr 28, 2005 (gmt 0)

So near as I can tell if you change the text of your ads the average CPC will go immediately up until the new ad's CTR history "overcomes" the old ad.

Which may take some time on an ad 6 months running that had built up the credibility to broad match more liberally.

Appears to be a disincentive to improve one's ad copy. So coming up with an improved, more relevant ad idea in the shower will cost you?

Too easy to be cynical about algo changes like this.....

patient2all

nyet




msg:1111870
 4:21 pm on Apr 28, 2005 (gmt 0)

But lets say you have scads of affliates setting up duplicate accounts on words they *know* have good CTR in order to get multiple ads on the same page.

How else to "weed" those out but start factoring ad CTR into the mix?

The more I think about this the clearer it becomes.

GerBot




msg:1111871
 5:08 pm on Apr 28, 2005 (gmt 0)

I would just like to know what the "teething period" is before a new ad will benefit from the CTR.
I'm happy with the new formula concept but what do I need to have my CTR factored in to my effective bid price?
do I need 1000 impressions?
do I need 100 clicks?
Do I need some other ratio of category Vs industry clicks over the total bid price of all advertisers minus the day parting average and the sum of the first number I think of?

nyet




msg:1111872
 5:20 pm on Apr 28, 2005 (gmt 0)

Before the particulars, I'd like even a confirmation that changing one's ad text can result in instantly higher CPC due to no ad CTR history?

This is a fundamental question one needs the answer to in order to proceed in logical fashion in managing one's own advertising budget.

I undertand the need for the secerecy of algo details, but the basics of how pricing works is not something I am happy with Google witholding.

Why can't this simple question receive an answer?

chrisk999




msg:1111873
 6:43 pm on Apr 28, 2005 (gmt 0)

Before the particulars, I'd like even a confirmation that changing one's ad text can result in instantly higher CPC due to no ad CTR history?

I tried to highlight this issue too in January [webmasterworld.com], with higher prices for new ads, to which AWA replied:

The short story is that, yes, the ads positioning algo has in fact shifted from considering only keyword CTR, and now folds-in information about the relevance of the ad itself, including the Display URL.

So yes, the Ad CTR is important, but the question is why the new ad's estimated CTR (for pricing) is NOT presumed to be the same as existing historical ads on the same keywords. It ought to be, but it definitely appears that it isn't, leading to inflated prices for new ads.

nyet




msg:1111874
 6:48 pm on Apr 28, 2005 (gmt 0)

The next unanswered question is where was the official notification that the basic pricing structure was changing? I can find none.

I have no problem with changing how you price things, but it should be disclosed prior to the change.

patient2all




msg:1111875
 12:14 am on Apr 29, 2005 (gmt 0)

Can you explain "adwords algorithm" to a dummy? Thanks

Zeus,

I don't think in the rancor anyone stopped to explain it to you. Well I'm kind of a dummy myself so I might be the best qualified :)

One of the features AdWords touts is that the high positions don't always go to the highest bidder as in Overture and other PPCs. In Overture, if I bid $1.00 and you bid $1.01 your ad is guaranteed to have a higher position than mine. Unless I bid $1.02 and now I'm #1, let's say.

AdWords claims to have a secret formula that makes the process fairer. In addition to the amount of your bid, your CTR (clicks/impressions) factor into how high you rank.

So potentially, one with a 5% CTR could bid 10 cents and rank higher than one with a 1% CTR who bids 50 cents. The theory being that the person with the 5% CTR has a better or at least more popular ad. So one with a high CTR is rewarded by paying less for a higher position than one with a lower CTR.

Now I'm not too good in math, but I think that makes the formula for who shows highest as (CTR * MaxCPC), at least in its simplest form.

Now it's being alleged that a third factor is suddenly being figured into the mix - AD CTR which you see next to your ad indicating how high the CTR is for your ad. When one makes any change to an ad, the ad in effect does not change, but becomes a new ad and starts with a 0 AD CTR. The ad must then prove itself again. This wasn't much of an issue until it became factored into your CPC (Cost Per Click). This makes it undesirable to change an ad since you're now brand new, in effect and lose much/some/a little of your bidding power.

When the equation was simply CTR * MaxPPC or whatever, as long as you held your CTR, you didn't have to worry about your bid fluctuating.

FWIW, I change ads all the time and have noticed no such effect. I stay out of highly competitive areas and sell several hundred items (like you I think) and always stay happily in one of the first 4 positions.

So many advertisers think customers are seeing ad #1 and thinking it must be the best. It's the one that catches the customer's eye that gets the business. Customers do not count down from the top to decide who to buy from. The sooner I got over that fallacy myself and concentrated on a catchy ad, the more success I had. Don't want to give away too many secrets to the sharks but sitting at the bottom of say 4-6 ads can make you stand out nicely. Couple that with an eye-catching, honest ad and you've got sales, inexpensive sales too!

----

Also, Nyet, I'm a little upset by your baseless suggestion:

lets say you have scads of affliates setting up duplicate accounts on words they *know* have good CTR in order to get multiple ads on the same page

Have any proof of this? I'd say it's more likely corporate entities would pull something like this since they can hide behind different divisions and have the financial resources to better pull it off and are the ones always engaged in bidding wars and other less than honorable tactics like impression spamming.

Most affiliates get good CTR like I have by creating relevant ads, staying out of bidding wars and concentrating on finding the most profitable items to market. All we need is one good ad and one good customer. I don't even worry about what the competition is doing. In fact for all my recent complaints about the irrelevancy of large corporate ads who bid on everything, I think it helps me. By the time they get to a refreshing, straightforward, folksy sounding ad that doesn't insult their intelligence, they're impressed.

Try looking elsewhere for the source of your problems instead of blindly blaming it on affiliates, Nyet.

Thanks,

patient2all

ogletree




msg:1111876
 4:02 am on Apr 29, 2005 (gmt 0)

I had a high up Adwords guy tell me at a conference that the programmers make changes without telling anybody. It is real easy to make a change but real hard to publish it. They have to go through a big process including lawyers before they publicly announce something. He said it is not uncommon for adwords advertisers to inform adwords reps of changes. There is very little comunication between customer service people and tech people at google.

nyet




msg:1111877
 4:21 am on Apr 29, 2005 (gmt 0)

P2a,

Baseless?

[webmasterworld.com...]

I figure doing this I may have two ads showing everytime a search is done on my keywords.

Sorry you are taking it personally P2A. I have no idea how much is going on. But I don't think G's slew of new rules aimed at affliate marketing are them just being "mean".

If there are people "signing up their wife's credit cards" to second accounts so they can have multiple ads on the same page it is reasonable to speculate that the new rules have something to do with preventing that behaviour.

Most affiliates get good CTR like I have by creating relevant ads, staying out of bidding wars and concentrating on finding the most profitable items to market.

Now I have to turn you question back on you. "do you have any proof for this?

As for whom to blame. I don't balme affliates for our losses due to the pricing/rank formula change. I just think perhaps they are related.

nyet




msg:1111878
 4:27 am on Apr 29, 2005 (gmt 0)

In fact, I blame Google for the loss. I have no problem with them changing anything they want about their business. It is their business. They could ban all affliate advertising or ban all non-affliate advertising -- it is their business anbd up to them how they constitute their offering.

The problem I have is the lack of clear (or any that I can find) disclosure that the basic and fundamental pricing model was changing.

patient2all




msg:1111879
 6:08 am on Apr 29, 2005 (gmt 0)

Aw 'cmon Nyet ;)

That thread was one guy's pipe dream. He didn't even say he did it. He was just thinking about it.

My sources that it's not happening?

A better sampling from among a circle of perhaps 25-30 affiliates that I deal with, no one has ever thought of the idea of a 2nd account.

Without even thinking it through, it doesn't even seem practical. Better to be diversified among 5-10 product lines than to try to attack one segment and waste 5-10 precious impressions for the keyword for one possible sale.

-----
I certainly agree with you that Google's complete lack of notification and communication with its advertisers on important policy changes is creating the impression that they could care less if they alienate a large number of their advertisers. This forum is hardly an official enough venue for policy changes to be announced.

They must realize that PPC is all they have as a true income base right now.

Their perceived arrogance toward paying advertisers will cost them once someone else refines a comparable, perhaps superior product.

Google is playing with animated ads when there is a house of cards blatantly evident in "Made for AdSense" empires whose fraudulant practices may account for up to 20% of their income.

The "shoplifting/just a cost of doing business" analogy is getting a little tiring. Couple that with the new TOS where the determination of click fraud will only be made by Google and then recompensed only with credits for more possible click fraud is really pushing the envelope.

patient2all

robertskelton




msg:1111880
 11:08 am on Apr 29, 2005 (gmt 0)

My big question is this:

If I change my ad every couple of days to reflect special offers, will I be at a disadvantage?

I mean that:

1) My keywords are unchanged
2) My ads do not create enough impressions to generate a "history" before they are updated

Would running new ads cost me more? Google actively encourages us to improve, improve, improve...

Ad CTR, with time, affects keyword CTR. Things were fine as they were. The new algo only has a tiny impact on things from a surfer's viewpoint, but a big impact on Google's earnings from those who regularly update or try to improve their ads.

And here's a thought - by letting us choose to not change our ads very often, Google's editorial staff will have less of a backlog!

robertskelton




msg:1111881
 11:15 am on Apr 29, 2005 (gmt 0)

BTW...

AWA says the algo has been in effect since Jan.

Any more recent increases in CPC could be be related to the new Budget Optimizer tool (other advertisers allowing Google to pump their max cpc) - especially if you hold a #1 position with a much higher max cpc than actual.

We have noticed this - especially where corporate advertisers have been using very vaguely related keywords.

Suddenly someone advertising pink widgets, previously at #7 for pink butterflies, is now at #1 or #2. And we are thinking - what idiots! They are costing us more to maintain position, but hey, it's a free world.

nyet




msg:1111882
 11:27 am on Apr 29, 2005 (gmt 0)

Any more recent increases in CPC could be be related to the new Budget Optimizer tool

not so. If you have not changed your ad since January you would not see the immediate CPC increase until your make the text change.

Dr_X




msg:1111883
 12:26 pm on Apr 29, 2005 (gmt 0)

not so. If you have not changed your ad since January you would not see the immediate CPC increase until your make the text change.

Nyet is right. There were no wild cost fluctuations on my ads until I made a very minor change on Feb 2nd. My cost went up by a factor of 4 just by removing an errant character in the ad. That's rediculous. Not knowing what was going on, I thought I had to further refine my ads. That only made things worse since costs were constantly fighting me. Imagine for a year you're paying "x" dollars per month, then suddenly you have to start paying 4 times that just because G makes a change to the rules without telling you. All last year I was able to make changes without worries. Now I'm afraid to touch anything (until some one points to a solution anyway).

-Dr.X

tenerifejim




msg:1111884
 1:24 pm on Apr 29, 2005 (gmt 0)

My suggestions for you guys who want to try new ads is to create a second or third one first in the ad group and run with them all for a while (turning off the "intelligent" Google selection option).

Then weed out the least sucessful options. If you end up with a newly created ad as your most successful then you will also have some "ad CTR history". I know this 'could' still increase cost, but I've found it an excellent way to work out (often suprisingly) what the best adverts are.

Now if only I could test advert conversion rates in the same way...

limitup




msg:1111885
 3:55 pm on Apr 29, 2005 (gmt 0)

I'm with Dr_X - same exact situation here. Bogus if you ask me. What's worse is that there are no clear answers from anyone about this ...

Hollywood




msg:1111886
 6:11 pm on Apr 29, 2005 (gmt 0)

1) Everything about this topic is confusing
2) The start of this topic started off with no real good explanation of what this thread was about in logical terms - or getting to the point fast of the problem & explaining the problem
3) Google will not jump in as they should
4) Some here are trying to explain - thanks people

That's all - just had to get that out...

Hollywood

AdWordsAdvisor




msg:1111887
 6:33 pm on Apr 29, 2005 (gmt 0)

1) Everything about this topic is confusing
2) The start of this topic started off with no real good explanation of what this thread was about in logical terms - or getting to the point fast of the problem & explaining the problem
3) Google will not jump in as they should
4) Some here are trying to explain - thanks people
That's all - just had to get that out...

Interesting points Hollywood.

Regarding #1, it does seem confusing, since this is the third thread that has addressed the topic recently (and the others are linked from this one). But the thread didn't really start here, and so, it's hard to follow. I can only agree.

Regarding #3, I've actually posted at least 5 times (between the three threads) to offer all the clarification I'm able to offer. I've also made it clear a number of times that I'll be passing feedback regarding this topic on the right folks here at AdWords, which I have done, and which I'll continue to do. You need have no doubt that the concerns expressed here have been heard by exactly the right teams.

Just out of curiosity, I spent a few minutes seeing how many people are actually posting in these threads. Turns out that about 2/3 of the posts have come from just three people, of which I am one. ;)

Anyway, just to set reasonable expectations, I have back-to-back meetings and tasks between noon and about 5:30, with tiny windows in-between. But after that, I'll try to return to this thread and see if I can address a key point or two. ;)

AWA

Sujan




msg:1111888
 6:43 pm on Apr 29, 2005 (gmt 0)

Just out of curiosity, I spent a few minutes seeing how many people are actually posting in these threads. Turns out that about 2/3 of the posts have come from just three people, of which I am one. ;)

They're doing good work, so I thought it wasn't neccessary to scream "me too" all the time. But be sure, it's interesting for most of us :)

- Jan

AdWordsAdvisor




msg:1111889
 12:32 am on Apr 30, 2005 (gmt 0)

OK, back again.

I'd like to first say that I have heard and understood all the feedback given on this topic to date. And, more importantly, I have routinely passed it on to the right people here at AdWords. So please rest assured that your voices have been heard.

As previously mentioned I'm not able to give lots of detail about the algos behind AdWords, for reasons that I hope will be obvious. However, I've added a few more details below, which I hope will answer the pressing questions that have been coming up repeatedly. Beyond this, though, please know in advance that I'll not be able to provide more detail. I'll have to limit myself to passing on your feedback. ;)

But because our new ads don't have a history of CTR then this must be why our CPC went up 20% overnight!

Yeah who knows what kind of history is needed. It could be a lot more than 1k impressions...

...I'd like even a confirmation that changing one's ad text can result in instantly higher CPC due to no ad CTR history?

Actually, this notion of 'history' does not apply here.

The positioning/pricing of new ads is effected by what amounts to a predicted CTR of the combination of the ad copy and the keyword that brings it up. This means that 'ad CTR' is effectively different per keyword, and it's a mistake to draw blanket conclusions about an automatic and instant change to a higher CPC if one simply changes ad copy.

So, no, new ads will not begin at 'zero' as has been reported here (or at least surmised). In fact, new ads could actually begin with a better ad/keyword CTR than the previous ad, if it's a better ad, and it's better targeted to the keywords that brings it up.

So, again, the better the connection between the ad and the keyword, the better the predicted CTR is likely to be, and the better it's position is likely to be for a given Max CPC.

Bottom line, what occurs when you create a new ad depends in part on how well the ad is composed, and particularly on how well it is composed and targeted in relationship to the keyword that brings it up.

Appears to be a disincentive to improve one's ad copy. So coming up with an improved, more relevant ad idea in the shower will cost you?

No, it can help you. But remember that it is also the targeting between the ad copy and the keyword that matters. So be sure that you're creating a new ad that it is highly related to the keywords that cause the ad to appear.

Hope that helps.

AWA

This 54 message thread spans 2 pages: 54 ( [1] 2 > >
Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google AdWords
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About
© Webmaster World 1996-2014 all rights reserved